The overarching goal of this long-term research project is to develop a multilevel model to elaborate the preventable causes of the major chronic diseases of working age adults - hypertension, diabetes, ischemic heart disease, asthma, musculoskeletal and COPD. A corollary is the development of similar models for disability, early retirement and premature mortality. Etiologic factors of a prior interest include: 1) Income, education, job status, ethnicity, gender, "lifestyle" risk factors, and chronic exposure to chemical and physical exposures at work, as well as psycho-social demands of the job, and 2) Factors associated with worksite location including regional health influences, access to and quality of health care, and workplace culture. The proposed multilevel models are designed to explain the heterogeneity in disease incidence, disability and death between worksite locations as a function of individual and job characteristics of workers at each location, as well as differences in the locations themselves. A particular ambition of the project is to identify those differential determinants of chronic disease that accrue during working life, such as cumulative exposures to occupational stressors, presumed in part responsible for the steep gradients in disease prevalence that burden the population after retirement. The empiric data for this comprehensive model will be derived from analyses of a prospective cohort study of the US workforce of Alcoa, a manufacturing company of some 40 thousand men and women geographically distributed at 54 major locations in 24 states. Linking administrative datasets on relevant exposures and health outcomes, supplemented by survey data on a sub-cohort of 5000 individuals at 20 locations, we will address four broad hypotheses: A. A strong SES gradient in the incidence of major chronic diseases, disability and premature death is operating in this geographically, demographically and occupationally diverse US workforce; Components of SES acting separately as well as together contribute to this gradient B. Job strain, shift-overtime, as well as physico-chemical aspects of the work environment are strong and remediable causal factors for the SES gradient; these job factors also contribute to and modify the impact of known social-behavioral risk factors such as smoking and obesity C. Geographic differences, including but not limited to differential utilization of quality health care, contribute to, and perhaps modify, the SES gradient D. Organizational characteristics comprise a set of potent contextual factors which modify the relationships between individual and job related risk factors and health outcomes.